Current strategies for drug encapsulation and controlled release typically use organic vehicles such as polymers, liposomes and micelles.                (a) Polymeric systems can be broadly classified as:        
Inert Matrix systems where the drug is trapped inside an inert, non-degradable polymer matrix, and its release controlled by diffusion through the porous network. In-vivo administration of such non-biodegradable polymeric particles is limited by the fact that the polymers will concentrate in intracellular “pockets” (e.g. lysosomes) or tissue, inducing severe overload of non-metabolised material. This limits their use to trans-dermal patches, etc. Another significant limitation is that the release is non-specific, since it is not activated by specific sites within the body. Finally, drug molecules exhibit intrinsically small diffusion coefficients within such matrices, limiting their broad application to potent drugs.
Reservoir systems, where the active ingredient's release is controlled by diffusion through an encapsulating membrane, hollow fibre, etc. The key limitations of these systems are their low mechanical strength and chemical resistance, since the controlling membrane is relatively fragile and easily fouled.
Chemical systems, in which the active molecules are dispersed inside a biodegradable matrix (e.g. polymers such as polyorthoesters and polyanhydrides). The release rate is preferably controlled by the heterogeneous (surface) dissolution/degradation of the matrix. This restricts the range of polymers that can be employed as matrices to bioerodible polymers such as poly(glycolic acid), poly(DL) lactic acid, poly(glycolic-colactic acid), poly caprolactone, polyhydroxy butyrate, and poly dioxianone.
Solvent-activated systems (hydrogels), in which the matrix swells in the presence of specific solute/solvent systems, with subsequent release of the encapsulated species. However, such polymers often swell too rapidly to provide therapeutically useful release rates, and the development of these systems is still in its infancy. In such controlled delivery systems, the delivery is controlled either by matrix structure (e.g. pore network tortuosity), particle size, overall drug loading or matrix solubility. A limitation of polymeric systems is that they can typically only exploit one, or at most two, of these features, and any changes in the drug usually necessitates reformulation of the matrix system. In contrast, an important feature of the present invention is that all of these features can be manipulated using the same underlying chemistry, which provides a more generic approach to designing controlled release matrices for specific applications.
Moreover, while there are many polymeric materials that have been identified as having potential for controlled drug release, relatively few have been approved for use in either human or veterinary pharmaceutical products.                (b) Liposomes are the most highly developed carrier system, but suffer from problems with in-vivo stability, aging and limited shelf life.        (c) The thermodynamic instability of micelles (which depends on temperature, concentration, solution speciation, etc) limits their applicability for controlling release. They also exhibit intrinsically low drug loading        (d) Bioceramics are used in bone-repair procedures (inert bioceramics, porous active ceramics that promote osteo-reconstruction). The inert bioceramics have purely mechanical applications, e.g. hip-joints (because of their low coefficient of friction)-typically Al2O3 or Y-TZP. The porous ceramics (typically hydroxy apatites) serve as structural bridges and “scaffolds” for bone formation. Bioactive glass provides an interfacial layer for tissues growth that resists substantial mechanical forces. Bioactive glasses have also been proposed as matrices for the controlled delivery of bioactive substances.        
Various patents have been issued to matrices prepared by sol-gel based processes, for example:
U.S. Pat. No. 5,591,453 (awarded Sep. 1, 1997) discloses the use of sol-gel silica matrices for the controlled release of biologically active molecules. The application cited was for osteo-reconstruction, and was restricted to large gel monoliths or granules (typically 0.5 to 5 mm). The release is controlled either by drug loading or varying the surface to volume ratio. Possible interactions between the matrix and drug were ignored. British Patent 1 590 574 (awarded Mar. 6, 1981) discloses the concept of incorporating biologically active components in a sol-gel matrix. Embodiment as substantially spherical particles in the size range from several microns to several millimetres was envisaged. It was noted that the rate of release of the biologically active component from the matrix would depend on a number of factors, including the pH of the medium, size of particles, and composition/porosity/structure/water content/hydrophilicity of the gel. The only example given was of spray-dried particles produced from bohemite sols, from which all of the imipramine initially encapsulated was released within five minutes. WO 9745367 (issued Apr. 12, 1997) discloses controllably dissolvable silica xerogels prepared via a sol-gel process, into which a biologically active agent is incorporated by impregnation into pre-sintered particles (1 to 500 μm) or disks. The release was controlled by varying the dimensions and chemical composition of the xerogels. WO 0050349 (issued 31 Aug. 2000) discloses controllably biodegradable silica fibres prepared via a sol-gel process, into which a biologically active agent is incorporated during synthesis of the fibre. The release was primarily controlled by varying the dissolution rate of the fibres.